[Wikipedia-l] Parsing TeX

[Toby Bartels]

JeLuF wrote for the most part:

>regarding "%": I think we don't need TeX-comments in our formulas. We could
>just replace % by \percent{} before passing the input over to TeX.

Sure, but that's just one example.  We can't get by without braces.

>regarding "simple style vs. complex formulas": I changed the parser of my
>proof-of-concept-code to first check  whether the formula is "simple", that
>is only contains: digits, letters (small+capital), +-*/=(), ^, _, {} 
>and blanks. If this is the case, the formula will be set as HTML using 
><sub> and <sup>.

And what do the braces represent?  TeX { and } or TeX \{ and \}?
Consistency demands the latter, but this will confuse newbies.

>I still use [[math: ... ]]-syntax, the code is prepared to also support [$ $], too.
>Before implementing this I wanted to wait for Lee's proposal for the future
>Wiki Markup Language.

I do think that having LaTeX abilities can be a great blessing.
But I think that when we use it, it needs to be clear that it's special,
something that (like tables, probably, no matter how we end up doing that)
people will want to read about on [[Wikipedia:How does one edit a page]]
before attempting to do them.  And that means that it just shouldn't be used
for simple expressions, which must be easily editable.

I still think that an inline/display division is the best solution.
This often corresponds to simple/complex,
and it has a clear visual cue that people can spot and grow used to.
But I'm waiting for Jan's reply to my last post;
he can often improve an idea ^_^.


-- Toby Bartels
   <toby+wikipedia-l at math.ucr.edu>